Open-source software is becoming a gradually more popular as a software development method; some of the most successful softwares are for example: the Linux operating system, Mozilla, Apache web server and openoffice.org. Open source softwares viewed by many as being very good in terms of their usage, reliability, performance and market share. Mostly open source software developer focus on functionality and different feature of the software; on the other hand they ignore the user centric design requirement. In this thesis the importance of the usability in Open source applications, such as Openoffice.org are described and how usability can be measured by assessing user performance such as, satisfaction, effectiveness and acceptability. Openoffice.org is a freely available office suite in different operating system and with different languages. The OpenOffice.org 3.0 feature set is similar to the feature set of Microsoft Office 2003. It has word processing, spreadsheet and presentation applications all together within a common suite. But still this application suite is not much familiar among common users that cause a major usability threat for usability studies. The basic purposed of this thesis was to find out the issues and users satisfaction regarding this Openoffice.org. Empirical method of usability assessment such as thinks aloud, Questionnaires and interviews were used. / Folkparkvagen 20 Lag 10 Ronneby 37240 Sweden Mobile number 0046-700183591

Online examinations are an integral component of many online learning environments and a high-stake process for students, teachers and educational institutions. They are the target of many security threats, including intrusion by hackers and collusion. Collu-sion happens when a student invites a third party to impersonate him/her in an online test, or to abet with the exam questions. This research proposed a profile-based chal-lenge question approach to create and consolidate a student's profile during the learning process, to be used for authentication in the examination process. The pro-posed method was investigated in six research studies using a usability test method and a risk-based security assessment method, in order to investigate usability attributes and security threats. The findings of the studies revealed that text-based questions are prone to usability issues such as ambiguity, syntactic variation, and spelling mistakes. The results of a usability analysis suggested that image-based questions are more usable than text-based questions (p < 0.01). The findings identified that dynamic profile questions are more efficient and effective than text-based and image-based questions (p < 0.01). Since text-based questions are associated with an individual's personal information, they are prone to being shared with impersonators. An increase in the numbers of chal-lenge questions being shared showed a significant linear trend (p < 0.01) and increased the success of an impersonation attack. An increase in the database size decreased the success of an impersonation attack with a significant linear trend (p < 0.01). The security analysis of dynamic profile questions revealed that an impersonation attack was not successful when a student shared credentials using email asynchronously. However, a similar attack was successful when a student and impersonator shared information in real time using mobile phones. The response time in this attack was significantly different when a genuine student responded to his challenge questions (p < 0.01). The security analysis revealed that the use of dynamic profile questions in a proctored exam can influence impersonation and abetting. This view was supported by online programme tutors in a focus group study.